The proposed four year study seeks to understand the nature of the relationship between drinking, drug use, and sexual behaviors that are risky in terms of HIV transmission. It also seeks to determine the best method for obtaining daily event data on these behaviors. The study employs a daily event methodology in which the temporal ordering of these effects can be ascertained, while at the same time this within-subjects design controls for stable individual differences that might account for the observed relationship between these factors. It also examines some hypothesized individual differences (e.g., propensity toward risk taking, misperceptions of risk) and situational factors (e.g., partner characteristics), which might account for or mediate the observed relationship between substance use and risky sex. The ultimate aim of the study is to provide needed empirical data to inform interventions and educational efforts aimed at reducing the incidence of AIDS. Four subpopulations known to engage in these risky behaviors will be studied: sexually active heterosexual adolescents, sexually active heterosexual adults (both clients of an inner city sexually transmitted disease clinic), sexually active heterosexual college students, and adult gay men. Equal numbers of males and females will be studied and the samples are expected to be racially and ethnically diverse. In addition to data on individual differences and situational factors, obtained in a baseline questionnaire, daily event data will be obtained on substance use, sexual activities, and situational factors surrounding these events. Within group and gender, subjects will be randomly assigned to report daily events either by keeping a diary or through daily telephone interviews. Data will be analyzed to assess possible causal relationships between substance use and risky sex, rule out competing hypotheses for the existence of this relationship, examine gender and group differences in the relationship, and compare the two methods for obtaining daily event data.